Item

Abstract

Responses of cortical neurons are highly variable. Even repeated presentations of the same visual stimulus never elicit the same spike train. Identifying the origins of this variability remains a challenge. There is increasing evidence that it is not just noise arising from stochastic features of neuronal architecture, but at least partly represents meaningful top-down signals. One of the most prominent examples of such top-down modulation in the visual system is covert attention. I will present both theoretical and experimental results showing that trial-totrial fluctuations of attentional state contribute significantly to response variability in primary visual cortex of awake, behaving monkeys. I will argue that much can be learned about information processing in the brain by using latent variable models of neuronal activity to help us identify and account for cognitive variables and make sense of single-trial neural population data.